Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [geomesa-users] GeoMesa 1.2.5 Released

Hi Alberto,

At the moment, our Accumulo and Kafka datastores have been the focus of our development and consequently those stores have the majority of features.  In terms of documentation, I think you are suggesting an area where we could communicate a little better.  At the moment, as features make it into the HBase datastore, more info will show up on http://www.geomesa.org/documentation/user/hbase_datastore.html.

As a silver lining, we have some work in progress to abstract over our indexing and query planning pieces.  That should make it easier to move large pieces of indexing support over to HBase.  Some of the pieces may make sense for Cassandra.  As always, improvements to open-source projects are supported either by direct funding or volunteer efforts; let me know if there's anything we can ever do to help make either of those easier!

Cheers,

Jim

On 08/15/2016 08:53 AM, Alberto Ramón wrote:

Hi Andrew
is there any doc about which features are for accumulo and which for HBase?

We tested Geomesa + hbase but we found hbase's support wasn't 100% functionality.  :(

Thanks. Alberto


El 15/8/2016 14:41, "Andrew Hulbert" <ahulbert@xxxxxxxx> escribió:
Hi everyone,

The next bugfix release of GeoMesa (1.2.5) is now available! Thanks to everyone who contributed bug fixes, new features, and ideas for improving GeoMesa. The full release notes can be found here: https://geomesa.atlassian.net/wiki/display/GEOMESA/GeoMesa+1.2.5+Release+Notes and are included below.

New versions of downstream projects including geomesa-ext, geomesa-nifi, and others will be updated in the next few days.

- Andrew Hulbert

---------------------------------

Highlights

* Improved Polygon and LineString Indexing

    - GeoMesa now includes XZ2 and XZ3 indexes for geometries with extents (e.g. non-points)
    - XZ indexes avoid object duplication and provide multi-level resolution indexing
    - XZ indexes are now the default for non-point geometries (e.g. LineStrings and Polygons)

* Configurable Date Intervals for Z3 and XZ3 Indexes

    - Z3 and XZ3 indexes can partition data by:
        - day/millis
        - week/seconds
        - month/seconds
        - year/minutes
    - Default indexing is week/seconds
    - Modifying the date intervals help with queries that are much larger or much smaller than a week

* Full Multi-Level Attribute Visibility Support

    - Updates partial support from 1.2.4
    - Attributes within a simple feature can have different visibility levels

* Jupyter Notebook Documentation

    - Documentation on using GeoMesa Spark with Jupyter Notebooks

Other Notable Changes

* Storage of FeatureIDs in Accumulo Keys

    - Decrease storage requirements for Accumulo schemas

* Command Line Tools & Distribution

    - List SFTs and Converters available to the environment
    - Improved installation & configuration scripts
    - Kafka 09 installation bundle now included in geomesa distribution tarball

* Bugs Fixes

    - Various documentation and build improvements.


Upgrading

    - GeoMesa 1.2.5 supports upgrade in place from versions 1.1.0.rc.7 through 1.2.4
    - As with previous versions, all jars in the system must match...Ensure that the geoserver jar, distributed runtime, and tools are all the same client version (1.2.5)
    - Shutdown the geoserver and upgrade the distributed runtime in accumulo as well as the geomesa jars in the webserver


_______________________________________________
geomesa-users mailing list
geomesa-users@xxxxxxxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://www.locationtech.org/mailman/listinfo/geomesa-users


_______________________________________________
geomesa-users mailing list
geomesa-users@xxxxxxxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://www.locationtech.org/mailman/listinfo/geomesa-users


Back to the top